Protocol for pro-inflammatory microRNA motif discovery using machine learning

Area of Science:

• Molecular Biology
• Bioinformatics
• Immunology

Background:

• MicroRNAs (miRNAs) play crucial roles in regulating inflammatory responses.
• Identifying specific miRNA sequences associated with pro-inflammatory activity is essential for understanding and manipulating immune responses.
• Current methods for miRNA functional analysis can be time-consuming and require extensive experimental screening.

Purpose of the Study:

• To present a novel protocol for identifying nucleotide motifs that predict the pro-inflammatory properties of miRNAs.
• To enable systematic motif discovery and biomarker prioritization directly from miRNA sequences.
• To streamline translational applications of miRNA research without extensive functional screening.

Main Methods:

• The protocol involves cell culture and miRNA transfection in macrophages.
• It integrates in vitro macrophage assays with k-mer discovery and motif searches.
• Least absolute shrinkage and selection operator (LASSO) regression is employed to identify predictive nucleotide sequence features.

Main Results:

• The workflow successfully identifies nucleotide motifs associated with pro-inflammatory miRNA activity.
• It demonstrates the ability to distinguish pro-inflammatory miRNAs based on their sequence features.
• The method provides a streamlined approach for biomarker prioritization.

Conclusions:

• This protocol offers a systematic and efficient method for discovering sequence-based biomarkers of miRNA pro-inflammatory function.
• It facilitates the translation of miRNA research into practical applications by reducing the need for extensive functional screening.
• The approach enhances our ability to predict and potentially modulate miRNA-driven inflammation through sequence analysis.